{"title":"Enhancing tourism demand forecasting with a transformer-based framework","authors":"Xin Li , Yechi Xu , Rob Law , Shouyang Wang","doi":"10.1016/j.annals.2024.103791","DOIUrl":null,"url":null,"abstract":"<div><p>This study introduces an innovative framework that harnesses the most recent transformer architecture to enhance tourism demand forecasting. The proposed transformer-based model integrates the tree-structured parzen estimator for hyperparameter optimization, a robust time series decomposition approach, and a temporal fusion transformer for multivariate time series prediction. Our novel approach initially employs the decomposition method to decompose the data series to effectively mitigate the influence of outliers. The temporal fusion transformer is subsequently utilized for forecasting, and its hyperparameters are meticulously fine-tuned by a Bayesian-based algorithm, culminating in a more efficient and precise model for tourism demand forecasting. Our model surpasses existing state-of-the-art methodologies in terms of forecasting accuracy and robustness.</p></div>","PeriodicalId":48452,"journal":{"name":"Annals of Tourism Research","volume":"107 ","pages":"Article 103791"},"PeriodicalIF":10.4000,"publicationDate":"2024-05-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Annals of Tourism Research","FirstCategoryId":"91","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S0160738324000689","RegionNum":1,"RegionCategory":"管理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"HOSPITALITY, LEISURE, SPORT & TOURISM","Score":null,"Total":0}
引用次数: 0
Abstract
This study introduces an innovative framework that harnesses the most recent transformer architecture to enhance tourism demand forecasting. The proposed transformer-based model integrates the tree-structured parzen estimator for hyperparameter optimization, a robust time series decomposition approach, and a temporal fusion transformer for multivariate time series prediction. Our novel approach initially employs the decomposition method to decompose the data series to effectively mitigate the influence of outliers. The temporal fusion transformer is subsequently utilized for forecasting, and its hyperparameters are meticulously fine-tuned by a Bayesian-based algorithm, culminating in a more efficient and precise model for tourism demand forecasting. Our model surpasses existing state-of-the-art methodologies in terms of forecasting accuracy and robustness.
期刊介绍:
The Annals of Tourism Research is a scholarly journal that focuses on academic perspectives related to tourism. The journal defines tourism as a global economic activity that involves travel behavior, management and marketing activities of service industries catering to consumer demand, the effects of tourism on communities, and policy and governance at local, national, and international levels. While the journal aims to strike a balance between theory and application, its primary focus is on developing theoretical constructs that bridge the gap between business and the social and behavioral sciences. The disciplinary areas covered in the journal include, but are not limited to, service industries management, marketing science, consumer marketing, decision-making and behavior, business ethics, economics and forecasting, environment, geography and development, education and knowledge development, political science and administration, consumer-focused psychology, and anthropology and sociology.